US 2026/0268284 A1Published application

Techniques for automatic tasks in an organizational management platform using a machine-learned model

OwnerNilay Laxmikant Pochhi, Vivin Flawence Peris +9inventors

InventorsNilay Laxmikant Pochhi, Vivin Flawence Peris, Kelly Michelle Trinh, Lakshya Poddar, Samarth Sharma, Sahadeva Hammari, Chad Tolentino, Prateek Chawla, Vikram Singh Deoelya, Ruhitaj Reddypalli, Colin King

Life of this patent

Estimating the term…

patentgrep

Loading patent…

Abstract

Techniques for automatically performing a task in an organizational management platform. For example, a computer system can include a database that collectively store the organizational data associated with the organization. The organizational data can include an object graph data structure comprising a plurality of data objects that respectively correspond to a plurality of entities of the organization. The system can receive, from a user device, a user query requesting a report associated with the organizational data. The system can determine, based on the user query, a first task from a plurality of tasks. The system can process the user query, the first task, and the organizational data to generate a prompt. The system can process, by a machine-learned model, the prompt and the organizational data to generate an output. The system can cause, on a display of a user device, a presentation of the output.

Claims22

Dependent claims

A computer system of an organizational management platform, the computer system comprising:

one or more processors;
and
one or more databases that collectively store organizational data associated with the organization, wherein the organizational data comprises an object graph data structure comprising a plurality of data objects that respectively correspond to a plurality of entities of the organization;
and
one or more non-transitory computer-readable media that collectively store instructions that, when executed by the one or more processors, cause the computer system to perform operations, the operations comprising:
receiving, from a user device, a user query to perform a task associated with the organizational data;
determining, based on the user query, a first task from a plurality of tasks;
determining, based on the first task, a clock-in action from a plurality of actions;
processing, the user query, the first task, and the organizational data to generate a prompt;
processing, using a machine-learning model, the prompt and the organizational data to generate an artificial intelligence (AI) generated output;
detecting a hallucinated element within the AI-generated output by comparing an AI-generated attribute of the AI-generated output with a previously generated report;
identifying, from a list of string choices in the computer system, a valid system string that represents a closest string match to the hallucinated element;
updating a parameter of the AI generated output based on the valid system string;
generating, based on the clock-in action and the updated parameter of the AI generated output, a first graphical user interface comprising a clock-in widget;
causing, on a display of a user device, a presentation of the AI generated output.

12 more elements

Dependencies are inferred from the claim text.